Dynamic-State-Estimation-Based Cyber Attack Detection for Inverter-Based Resources

The cyber security of Inverter Based Resources (IBRs) has received increasing attention in recent years. In this paper, a cyber attack detection method is proposed based on dynamic state estimation for IBRs. The state-space models of the physical inverter system and the digital controller are separately derived and then coupled by the data flows in between, i.e., measurement signals and control signals. Based on Kalman Filtering (KF), two dual dynamic state estimators are developed for tracking the cyber and physical state variables, respectively. By checking the model-data consistency of both the physical and cyber layers via hypothesis testing, the proposed method features the capability of distinguishing between false data in measurement signals and in control signals. Simulation results in an IEEE 13 node test feeder demonstrate the effectiveness of the proposed method.